DUOLINGO COULDN'T TEACH ME FRENCH IN 2 YEARS

ChatGPT did it in just 4 weeks.

Here are the 8 exact prompts I used ↓ 1. Creator of daily lessons

«Act as a French teacher. Create a daily 20-minute lesson with vocabulary, grammar, and practical exercises»

1. Open Gemini
2. Upload your photo
3. Enter this prompt
4. Generate
5. Change the pose

Here are 10 best prompts (PART 4)🧵
Prompt 1:

(this prompt works for both genders)

'Create a black and white high-fashion editorial portrait of me, keeping my real face exactly as in the reference photo (no edits, same features, no retouching). I am wearing a loose white blouse. My pose is bold and expressive, with one arm raised to hold my hair up, partially covering my face. My gaze is intense and directed at the camera, creating a strong and enigmatic aura. The background is a plain minimalist wall with sharp natural daylight casting defined shadows, adding depth and contrast.
Lighting is harsh and high-contrast, highlighting my facial contours, sharp cheekbones, and the texture of the hair and fabric. The style should feel raw, dramatic, and artistic.
The overall image is high-fashion editorial in monochrome, cinematic, and striking.'

My friend was unemployed for 8 months.

I rewrote his resume using ChatGPT. 28 days later, he got hired.

Here are 8 prompts that actually worked ↓ 1/ Resume Optimizer

"This is my resume: [paste text]. Rewrite it for [job position]. Highlight quantifiable results and relevant skills, and adapt the format for recruiters and applicant tracking systems (ATS)"

I stopped buying $500 courses when I realized ChatGPT teaches better.

Here are 9 prompts that turned me into my own teacher: 1/ Prompt:

“Explain [topic] to me like I’m 12, then give me the adult version after I understand it.”

I just dropped a step-by-step roadmap to build your first AI agent and it’s pure gold.

Here’s how it actually works: 🧵 1/ I made an agent that writes viral-style tweets based on my style and topics.

it understands tone, flow, and structure, and even suggests hooks that sound like me.

took under 3 mins to make.

1. Open Gemini
2. Upload your photo
3. Enter this prompt
4. Generate
5. Change the pose

Here are 10 best prompts (PART 3)🧵
1/ Prompt:

Ultra-realistic studio portrait of a woman leaning gracefully on the backrest of a chair, dressed in a sleek dark outfit. Shot from a subtle low-angle to accentuate her facial features and confident expression. Cinematic black-and-white color grade with refined, natural skin tones. Solid black background with sharp 8K details, highlighting textures in her clothing, hair, and eyes for a dramatic, lifelike effect.

1. Open Gemini
2. Upload your photo
3. Enter this prompt
4. Generate
5. Change the pose

Here are 10 best prompts (PART 2)🧵

1/ Prompt:

A 4K ultra-realistic portrait of a young girl, framed from the chest up (not full body), designed for Instagram story or profile picture. The background is a luxurious Persian red carpet design (only behind her, not on the ground). Sunlight shines from one side,casting natural warm light across half of her face. She is wearing an elegant dress inspired by Persian carpet patterns,with authentic Iranian colors and motifs. Her facial features remain unchanged.She is holding a Persian cat in her hands

Nano-Banana is everywhere🍌

Here’s the ultimate resource to master its creative editing power

20+ Nano-Banana use cases, each with the prompt, input, and output side by side: (Part 2)
Case 1: OOTD Outfit（by @302.AI）

input: Need to upload a person image and clothing image

PROMPT:
Choose the person in Image 1 and dress them in all the clothing and accessories from Image 2. Shoot a series of realistic OOTD-style photos outdoors, using natural lighting, a stylish street style, and clear full-body shots. Keep the person's identity and pose from Image 1, but show the complete outfit and accessories from Image 2 in a cohesive, stylish way.

this is crazy!

The line between real and fake is gone and people are loving it.

Here are 10 viral clips where AI meets absolute madness:

1. no one can tell it's AI anymore... 2. Smart kitty

1. Open Gemini
2. Upload your photo
3. Enter this prompt
4. Generate
5. Change the pose

Here are 10 best prompts (PART 1)🧵

2. Prompt:

A high-resolution black and white portrait of a girl wearing a sharp black turtleneck and chic small gold earrings, along with a Korean bun. She stands confidently in a minimal studio setting, slightly to the side. Dramatic studio lighting creates bold geometric shadows on her face and background with strong cross beams of light. Cutting through the shadows, her expression is calm, introspective, and slightly distant. The image has a cinematic atmosphere.

The PowerPoint era is over.

You don’t need 35-year-old tech to make slides anymore.

Here's how presentations can be created in 2 minutes by entering a prompt like in ChatGPT: We’ll be using @abacusai DeepAgent.

It does an excellent job with PowerPoint presentations and deep research.

Check it out now: deepagent.abacus.ai

Nano-Banana is everywhere🍌

Here’s the ultimate resource to master its creative editing power

60+ Nano-Banana use cases, each with the prompt, input, and output side by side: (Part 1) Case 1: Illustration to Figure（by @ZHO_ZHO_ZHO）

input: Need to upload a reference image as the object for generating the figure

PROMPT → 'turn this photo into a character figure. Behind it, place a box with the character's image printed on it, and a computer showing the Blender modeling process on its screen. In front of the box, add a round plastic base with the character figure standing on it. set the scene indoors if possible'

OpenAI has released a shocking paper that reveals

"why AI hallucinates, and its mechanism."

And it's free,

Here's are true reasons why hallucinations occur, as shown in the paper, along with 6 solutions🧵

1. AI is trained in such a way that it cannot say

"I don't know"

The biggest cause of hallucinations lies in the AI's training method itself.

In the current evaluation systems, even if the answer is incorrect, guessing provides a higher score than answering "I don't know," so the AI ends up learning to actively lie (bluff).

AI just ended PowerPoint.

Forget slides, effort, planning, and endless formatting.

5 AI tools that remove PowerPoint from existence: 1. Plus AI

Turn ideas into high-quality slides in seconds.

Works right inside PowerPoint and Google Slides, fully compatible and seamless.

plusai.com/ai-powerpoint-…

The most important skill you’ve never been taught:

Prompting.

10 dead-simple tips to get better results from AI (worth saving): 1. Cut the fluff

Most prompts are vague, rambly, or full of filler.
AI doesn’t need your backstory it needs clarity.

✗ “Hey can you maybe explain AI a little bit, like in a way someone could understand?”

→ “Explain AI in 3 bullet points as if I’m 12 years old.”

The new Nano banana is like a Photoshop on steroids.

It lets you do INSANE things that weren't possible before.

Here are the 20 coolest examples I've found 🧵 1/ Combine photos into new scenes

Andrew Ng has been right about AI every single time.

Now, he’s made his boldest prediction yet:

“5 upcoming AI opportunities that could create more millionaires than anything before.”

Here’s what you need to know: 🧵 First, his track record:

He built Google Brain, co-founded Coursera (over 120M learners), and led AI at Baidu, managing a team of 1,300 researchers.

He’s trained 8M+ students and runs an AI fund worth $370M.

When he talks about the future of AI, people listen.

🚨 Apple is in trouble.

Google just dropped the Pixel 10… and it feels like it’s from 2030.

The iPhone suddenly looks outdated.

Here are 10 wild AI features that change everything: 1. In the blink of AI, turn ordinary shots into extraordinary moments with a world-class camera system.

Amazon just dropped Kiro.

It not only writes code but also creates clear specifications, generates and executes tasks, and even detects bugs.

Here are some wild examples (and yes-it’s FREE): 1/ Convert prompts into requirements definitions From instructions like "Add a review function," generate clear requirements that developers can immediately act upon.

It's been an insane week for AI.

1. Google Doppl
2. HeyGen launches Video Agent
3. Google releases Gemini CLI
4. AlphaGenome by DeepMind
5. OpenAI loses key researchers to Meta
6. ElevenLabs launches 11ai VA
7. Tencent’s GameCraft and more...

Everything you need to know: 🧵 1. Google launched Doppl, a virtual try-on app for clothes.

Just upload your photo, and it shows you wearing different outfits, even keeping your accessories like bags or shoes.

It also animates your look to make it more realistic.

Currently available in the US (18+).

The global tech race is shifting.

China is rapidly outpacing the US in robotics — real machines, real deployment.

Software is no longer enough.

Here’s how China took the lead — 12 examples: 🧵
1. Unitree H1 Humanoid

A $90K humanoid robot that can walk, run, and balance, at 5% of the US equivalent’s cost.

This isn’t from a megacorp. It’s from a startup in Shenzhen.

The secret? China’s mature supply chains and rapid iteration cycles.